{"ID":6536382,"CreatedAt":"2026-07-14T01:21:01.169441415Z","UpdatedAt":"2026-07-14T08:17:36.581265645Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.10203","arxiv_id":"2607.10203","title":"When Does Depth Survive Composition? Compute--Quality Regimes in Latent World Models","abstract":"Adaptive-compute world models -- early-exit or mixture-of-depths predictors that spend variable depth per step -- assume depth buys better predictions and can be routed adaptively. In autoregressive rollouts, the first assumption requires depth's per-step precision to survive composition. We test this with a pre-registered instrument, the shallow penalty $ρ=\\mathrm{err}(\\text{shallowest-exit rollout})/\\mathrm{err}(\\text{full-depth rollout})$, across nine DeepMind Control tasks under matched single-step ($K=1$) and multi-step ($K=4$) training, three seeds each. We find three regimes: on 6/9 tasks depth helps rollouts (intrinsic, $ρ$ up to $4.7\\times$), on 2/9 the shallow exits beat the full stack (inversion, $ρ$ down to $0.85\\times$), and one is flat. The robust inversion (cheetah) is not a property of the dynamics but is created by training: an ablation supervising early exits only at the first rollout step erases it ($ρ: 0.87\\to1.18$, $n=8$, $Δ=+0.31$), while an intrinsic-tradeoff task is unaffected -- a double dissociation we call the routability catch-22, since the supervision that makes exits routable is what trains them to out-roll the full stack. The regime is partly predictable a priori: observation/action dimensionality and one-step model error correlate with $ρ$ at $|\\text{Spearman}|\\approx0.75$ ($n=9$). Inside a CEM planner, $ρ$'s sign predicts whether planning benefits from depth, most sharply on the inversion task, where shallow planning beats deep. Finally, three cautions: a task's regime depends on the metric space, the rollout horizon, and the encoder. All thresholds and gates were fixed before the compute campaign, including a pre-registered negative for the hypothesis that motivated the study.","short_abstract":"Adaptive-compute world models -- early-exit or mixture-of-depths predictors that spend variable depth per step -- assume depth buys better predictions and can be routed adaptively. In autoregressive rollouts, the first assumption requires depth's per-step precision to survive composition. We test this with a pre-regist...","url_abs":"https://arxiv.org/abs/2607.10203","url_pdf":"https://arxiv.org/pdf/2607.10203v1","authors":"[\"Achyuthan Sivasankar\"]","published":"2026-07-11T08:27:33Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[]","has_code":false}